| action-recognition-in-videos-on-kinetics-400 | Florence | #1 | Top-1 Accuracy: 86.5Top-5 Accuracy: 97.3 |
| visual-question-answering-on-vqa-v2-test-std-1 | Florence | #3 | overall: 80.36 |
| visual-question-answering-on-vqa-v2-test-dev-1 | Florence | #7 | Accuracy: 80.16 |
| cross-modal-retrieval-on-coco-2014 | Florence | #9 | Text-to-image R@1: 63.2Text-to-image R@5: 85.7Image-to-text R@1: 81.8… |
| image-classification-on-imagenet | Florence-CoSwin-H | #11 | Top 1 Accuracy: 90.05%Number of params: 893MTop 5 Accuracy: 99.02 |
| zero-shot-cross-modal-retrieval-on-flickr30k | Florence | #11 | Image-to-text R@1: 90.9Image-to-text R@5: 99.1Text-to-image R@1: 76.7… |
| zero-shot-cross-modal-retrieval-on-coco-2014 | Florence | #18 | Image-to-text R@1: 64.7Image-to-text R@5: 85.9Text-to-image R@1: 47.2… |
| zero-shot-video-retrieval-on-msr-vtt | Florence | #19 | text-to-video R@1: 37.6text-to-video R@5: 63.8… |
| action-classification-on-kinetics-600 | Florence (curated FLD-900M pretrain) | #20 | Top-1 Accuracy: 87.8Top-5 Accuracy: 97.9 |
| video-retrieval-on-msr-vtt-1ka | Florence | #40 | text-to-video R@1: 37.6text-to-video R@5: 63.8… |
| object-detection-on-coco-test-dev | Florence-CoSwin-H | – | box mAP: 62.4 |
| object-detection-on-coco-val2017 | Florence-CoSwin-H | – | box AP: 62 |